Oratis commited on
Commit
63acf9b
·
verified ·
1 Parent(s): f42abf8

Document reviewed gameplay archive, schema and original quality limitations

Browse files
Files changed (2) hide show
  1. NOTICE.md +7 -0
  2. README.md +55 -0
NOTICE.md ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ # Upstream attribution and archive terms
2
+
3
+ Scenes and assets are from the CARLA simulator (https://carla.org), especially Town10HD_Opt / CARLA 0.9.15. CARLA's official README identifies MIT for its code and CC-BY for CARLA-specific assets; it does not specify a CC-BY version there, so this archive does not invent one. Consult the upstream notices and any third-party components: https://github.com/carla-simulator/carla/blob/0.9.15/README.md#licenses.
4
+
5
+ The Diogenes project owner authorizes public archival of its own captures, generated camera plans and technical annotations. Upstream engine, asset and third-party rights remain with their respective owners. This publication does not relicense simulator source assets or claim new unrestricted commercial/ML-training rights. No CARLA executable or source-asset package is included.
6
+
7
+ Synthetic ambience, generated camera/input fields, historical failed checks and unresolved conventions must not be represented as real recorded acoustic/action ground truth or certified production data. The separately published Diogenes source-code licence does not automatically replace upstream asset terms.
README.md ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ pretty_name: Diogenes CARLA Capture Iterations Archive v01
3
+ license: other
4
+ license_name: attributed-simulator-capture-archive
5
+ license_link: https://huggingface.co/datasets/DiogenesLab/Diogenes_CARLA_Capture_Iterations_Archive_v01/blob/main/NOTICE.md
6
+ tags:
7
+ - video
8
+ - simulation
9
+ - robotics
10
+ - multi-view
11
+ - archival
12
+ size_categories:
13
+ - n<1K
14
+ ---
15
+
16
+ # Diogenes CARLA Capture Iterations Archive v01
17
+
18
+ Historical CARLA camera-capture iterations: **260 MP4 files, 755 JSON metadata files, 187 PNG inspection frames/contact sheets and 20 empty historical status markers**. The archive preserves **1,222 original objects / 24,090,799,445 bytes**, across nine collections. Versions can overlap; video-file counts do not imply independent scenes or unique runs.
19
+
20
+ | Collection | MP4 files | All objects | Bytes |
21
+ | --- | ---: | ---: | ---: |
22
+ | carla-final | 24 | 108 | 2,196,036,361 |
23
+ | carla-hz | 24 | 132 | 2,293,821,926 |
24
+ | carla-hz2 | 16 | 88 | 1,512,393,351 |
25
+ | carla-road | 20 | 112 | 1,903,560,577 |
26
+ | carla-v1 | 40 | 130 | 3,557,444,427 |
27
+ | carla-v2 | 40 | 170 | 3,750,163,949 |
28
+ | carla-v3 | 40 | 160 | 3,667,947,562 |
29
+ | carla-v5 | 28 | 161 | 2,630,682,431 |
30
+ | carla-v6 | 28 | 161 | 2,578,748,861 |
31
+
32
+ ## Layout and interpretation
33
+
34
+ Tar shards preserve relative paths under `collections/<collection>/`. Video groups generally have four camera lanes `video_1.mp4` through `video_4.mp4`, alongside `action_camera_N.json`, `systeminfo_N.json`, planning/provenance and available co-visibility/dynamic-object metrics. PNGs under `sheets`, `ov` or similarly named paths are original inspection views, not extra camera recordings. Some collections place additional planning files in subdirectories; the member manifest gives the exact layout.
35
+
36
+ Per-frame arrays contain time/frame/fps, virtual camera positions, rotations, intrinsics and speed, with version-dependent simulated player/input fields. The historical key `camera_Follow Offset` is intentionally unchanged. Some `time` fields are date-time strings; others are numeric seconds serialized as strings. Read each file's `field_conventions`: early planned metadata and later CARLA-native playback do not necessarily share an axis/rotation convention. Do not infer real human actions from synthesized input fields.
37
+
38
+ All representative videos are 60 seconds at nominal 30 fps. Earlier 4:3 views (1440×1080) coexist with 1920×1080 views. Existing media are not resized, trimmed or re-encoded. Some earlier videos have no audio; matching project tooling uses **synthetic brown-noise ambience** for muxed CARLA audio, not captured game-world sound. Treat that audio as synthetic and preserve the original per-file stream differences.
39
+
40
+ ```python
41
+ import json, tarfile
42
+ with tarfile.open("downloaded-shard.tar", "r:") as archive:
43
+ for member in archive:
44
+ if member.isfile() and member.name.endswith("provenance.json"):
45
+ metadata = json.load(archive.extractfile(member))
46
+ print(member.name, metadata.get("scene"), metadata.get("field_conventions"))
47
+ ```
48
+
49
+ ## Quality and review
50
+
51
+ This archive intentionally includes failed and intermediate experiments: obstructed camera views, extreme or incorrect orientation, near-object clipping and differences in dynamic-object visibility. Original self-checks and source notes remain intact. The 20 empty `.state-QC_PASS` files are historical automation markers; they are **not a new production certification or blanket evidence that all delivery requirements passed**. Original unresolved coordinate-convention caveats and synthetic-scene flags remain unchanged. No train/test split, uniform action coverage or new training eligibility is claimed.
52
+
53
+ All 755 JSON objects were fully parsed and scanned. Sensitive-field review resolved 30 `attribution` hits as the upstream CARLA attribution text, which is preserved. Exact-SHA comparison reused the already fully reviewed camera-plan payloads where applicable; additional variant field names/types and timestamp formats were inspected from their originals. Representative media inspection covered 72 video first frames and 14 PNG frames/contact sheets across all nine collections. No private account or desktop overlay was identified in those samples. This is not exhaustive video viewing or audio transcription.
54
+
55
+ One 6,148-byte macOS `.DS_Store` record is retained privately and excluded from public capture data. Every other listed capture object, including empty status markers and defective iterations, is preserved byte-for-byte. Publication proceeds incrementally; each shard's actual size/SHA-256 is checked against pinned Hugging Face LFS before the matching source is removed. Source tools and known historical caveats: [Diogenes Archive](https://github.com/CuddlerAI/Diogenes-Archive/tree/main/engine-data/carla-capture).